Artificial Intelligence and Data Literacy Vocabulary Flashcards

Comprehensive vocabulary flashcards covering the Artificial Intelligence domains, AI project cycle, data acquisition, visualization, modeling, evaluation metrics, AI ethics, and foundational math principles.

Statistical Data

Information mostly in the form of numbers or tables that can be analyzed and interpreted using statistical methods to help understand patterns, trends, and relationships.

Data Feature

An individual measurable property or characteristic of a data object being recorded or stored.

Surveys

A data collection method used to gather information from a set group of people about a wide range of topics.

Web Scraping

The process of collecting organized data from websites using software tools.

Application Program Interface (API)

A programming language interface that enables two apps or computer applications to communicate with one another.

Data Visualization

The graphical presentation of data using visual elements like charts, graphs, and maps to make complex data easy to understand, analyze, and communicate.

Artificial Intelligence (AI)

The development of computer systems capable of performing tasks that previously required human intelligence, such as speech recognition, decision-making, and pattern identification.

Machine Learning

The study of computer systems that learn and adapt automatically from experience without being directly programmed.

Deep Learning

An advanced AI technique composed of algorithms that permit software to train itself to perform tasks using massive volumes of unstructured big data without human intervention.

Evaluation

A systematic procedure that assesses the performance of an AI model by providing a test dataset as input and comparing predictions with actual outcomes.

Overfitting

A condition where an AI model remembers the entire training dataset, causing it to always predict the proper label for training points but failing to evaluate well on new data.

AI Bias

A phenomenon where an algorithm produces a learned output containing content systematically prejudiced toward a certain religion, race, caste, gender, or community.

Computer Vision (CV)

The domain of Artificial Intelligence used to read, process, and analyze visual data in the form of images and videos.

Natural Language Processing (NLP)

The domain of Artificial Intelligence that allows a machine to read, understand, and analyze meaning from human language using textual and speech data.

Natural Language Understanding (NLU)

A sub-part of NLP that helps machines process and understand user input provided in natural human languages like text or speech.

Natural Language Generation (NLG)

A sub-component of NLP that helps generate a machine response in a natural language based on the input provided by the user.

Problem Scoping

The process of understanding a problem, identifying its influencing factors, and defining the main objective of an AI project.

Stakeholders

People who are affected by an issue and would benefit from the solution created for it.

Structured Data

Data available in a standardized, well-defined format following a consistent order that is easily accessible by humans and programs, such as numbers and characters.

Unstructured Data

Data that deviates from standardized formats and traditional data models, such as video, audio, image, and log files.

Semi Structured Data

Information that does not fit neatly into a relational database but still retains some organizational structure to save space and provide flexibility, such as XML data.

Dataset

A collection of data in tabular format containing numbers or values related to a specific subject.

Training Dataset

A large portion of a dataset (typically 80%) used to teach a machine learning algorithm to make judgments or perform tasks.

Test Dataset

A portion of a dataset (typically 20%) identified to evaluate and test the performance of a trained computer program or model.

Rule-Based AI

An AI modeling approach where the machine operates on predefined rules and labeled data explicitly provided by the developer.

Learning-Based AI

An AI modeling approach where the machine learns patterns and trends on its own from random, unlabelled data fed into the system.

Decision Tree

A classification and prediction tool structured like a tree with a root node at the top and branching decision nodes ending in leaf nodes.

Prediction

The output or predicted outcome generated by an AI model when evaluated.

Reality

The actual or correct outcome existing in real-world data against which an AI model's prediction is compared.

True Positive (TP)

A outcome condition where both the predicted value of the AI model and the actual real-world value are positive.

True Negative (TN)

An outcome condition where both the predicted value of the AI model and the actual real-world value are negative.

False Positive (FP)

Also referred to as Type I Error, a condition where the predicted value of an AI model is positive, but the actual real-world value is negative.

False Negative (FN)

Also referred to as Type II Error, a condition where the predicted value of an AI model is negative, but the actual real-world value is positive.

Confusion Matrix

A chart or table that summarizes the effectiveness of a classification-based AI model by displaying predicted values alongside actual outcome values.

Accuracy

The percentage of correct predictions out of all total observations, calculated as $\text{Accuracy} = \frac{TP + TN}{TP + TN + FP + FN} \times 100\%$.

Precision

The rate at which desirable positive predictions turn out to be correct, calculated as $\text{Precision} = \frac{TP}{TP + FP} \times 100\%$.

Recall

The measurement of correct positive predictions relative to the total number of actual positive instances in a dataset, calculated as $\text{Recall} = \frac{TP}{TP + FN} \times 100\%$.

F1 Score

A metric measuring the balance between precision and recall, calculated as $\text{F1 Score} = 2 \times \frac{\text{Precision} \times \text{Recall}}{\text{Precision} + \text{Recall}}$.

Morals

An individual's or society's personal guidelines regarding what is right or wrong in human behavior.

Ethics

The logical and structured system of moral values determined by a community or group to assess fair and unfair actions.

Data Literacy

The ability to understand, interpret, work with, analyze, and communicate effectively with data.

Data Discovery

The initial data acquisition step focused on searching for valuable information across sources and checking its quality.

Data Augmentation

The process of increasing data amount and diversity by transforming existing data using modifications like flipping, rotating, or altering brightness.

Data Generation

The process of collecting or recording brand-new data directly using hardware devices or sensors.

Data Privacy

The management and governance of how personal data is collected, shared, and used, providing individuals control over access to their personal information.

Data Security

The practice of protecting digital information from unauthorized access, corruption, or theft throughout its entire lifecycle.

Cyber Security

The protection of computers, servers, mobile devices, electronic systems, networks, and data from malicious attacks.

Continuous Data

A category of numeric quantitative data that can take any fractional value along a continuous scale, such as height, weight, or voltage.

Discrete Data

A category of numeric quantitative data containing only whole numbers that cannot be split into fractions, such as the total count of students.

Probability

A branch of statistics measuring the likelihood or chance of an event occurring, expressed as $P(A) = \frac{\text{Number of Favourable Outcomes}}{\text{Total Number of Possible Outcomes}}$.